US20150347500A1 - Interactive searching method and apparatus - Google Patents

Interactive searching method and apparatus Download PDF

Info

Publication number
US20150347500A1
US20150347500A1 US14/572,393 US201414572393A US2015347500A1 US 20150347500 A1 US20150347500 A1 US 20150347500A1 US 201414572393 A US201414572393 A US 201414572393A US 2015347500 A1 US2015347500 A1 US 2015347500A1
Authority
US
United States
Prior art keywords
query
word
words
key
feedback
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Abandoned
Application number
US14/572,393
Other languages
English (en)
Inventor
Tingting Li
Wei Wan
Shiqi Zhao
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Baidu Online Network Technology Beijing Co Ltd
Original Assignee
Baidu Online Network Technology Beijing Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Baidu Online Network Technology Beijing Co Ltd filed Critical Baidu Online Network Technology Beijing Co Ltd
Assigned to BAIDU ONLINE NETWORK TECHNOLOGY (BEIJING) CO., LTD. reassignment BAIDU ONLINE NETWORK TECHNOLOGY (BEIJING) CO., LTD. ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS). Assignors: LI, TINGTING, WAN, WEI, ZHAO, SHIQI
Publication of US20150347500A1 publication Critical patent/US20150347500A1/en
Abandoned legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/903—Querying
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/24—Querying
    • G06F16/242—Query formulation
    • G06F16/2423—Interactive query statement specification based on a database schema
    • G06F17/30392—
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/24—Querying
    • G06F16/245—Query processing
    • G06F16/2457—Query processing with adaptation to user needs
    • G06F16/24578—Query processing with adaptation to user needs using ranking
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/95—Retrieval from the web
    • G06F16/951—Indexing; Web crawling techniques
    • G06F17/3053—
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
    • G06F3/048—Interaction techniques based on graphical user interfaces [GUI]
    • G06F3/0484—Interaction techniques based on graphical user interfaces [GUI] for the control of specific functions or operations, e.g. selecting or manipulating an object, an image or a displayed text element, setting a parameter value or selecting a range
    • G06F3/04842—Selection of displayed objects or displayed text elements

Definitions

  • Embodiments of the present disclosure generally relate to a searching technology field, and more particularly, to an interactive searching method and apparatus
  • the search engine obtains a search result associated with the query and returns the search result to the client and the user obtains the source that he needs from the returned search result finally.
  • Embodiments of the present disclosure seek to solve at least one of the problems existing in the related art to at least some extent.
  • a first objective of the present disclosure is to provide an interactive searching method, which can update a query automatically according to a historical query and a feedback, thus reducing an input operation of a user and decreasing a memory burden of the user.
  • a second objective of the present disclosure is to provide an interactive searching apparatus.
  • embodiments of a first aspect of the present disclosure provides an interactive searching method, including: receiving a first query; obtaining an intention clarification guidance sentence according to the first query; receiving a feedback corresponding to the intention clarification guidance sentence and generating a second query according to the first query, the intention clarification guidance sentence and the feedback; and providing a search result according to the second query.
  • Embodiments of a second aspect of the present disclosure provide an interactive searching apparatus, including: a first receiving module configured to receive a first query; a first obtaining module configured to obtain an intention clarification guidance sentence according to the first query; and a second receiving module configured to receive a feedback corresponding to the intention clarification guidance sentence; a generating module configured to generate a second query according to the first query, the intention clarification guidance sentence and the feedback; and a providing module configured to provide a search result according to the second query.
  • Embodiments of a third aspect of the present disclosure provide an apparatus, including: one or more processors; a memory; and one or more programs stored in the memory and executed by the one or more processors to execute steps of: receiving a first query; obtaining an intention clarification guidance sentence according to the first query; receiving a feedback corresponding to the intention clarification guidance sentence and generating a second query according to the first query, the intention clarification guidance sentence and the feedback; and providing a search result according to the second query.
  • Embodiments of a fourth aspect of the present disclosure provide a non-transitory computer-readable storage medium, including one or more programs for executing steps of: receiving a first query; obtaining an intention clarification guidance sentence according to the first query; receiving a feedback corresponding to the intention clarification guidance sentence and generating a second query according to the first query, the intention clarification guidance sentence and the feedback; and providing a search result according to the second query.
  • FIG. 1 is a flow chart of an interactive searching method according to an embodiment of the present disclosure
  • FIG. 2 a is a schematic diagram showing an effect of inputting a feedback according to an intention clarification guidance sentence according to an embodiment of the present disclosure
  • FIG. 2 b is a schematic diagram showing an effect of providing a search result according to a second query according to an embodiment of the present disclosure
  • FIG. 2 c is a schematic diagram showing an effect of providing a candidate result according to a first query according to an embodiment of the present disclosure
  • FIG. 2 d is a schematic diagram showing an effect of providing a search result according to a second query according to an embodiment of the present disclosure
  • FIG. 3 is a flow chart of a method for generating a second query according to a first query, an intention clarification guidance sentence and a feedback according to an embodiment of the present disclosure
  • FIG. 4 is a flow chart of a method for obtaining one or more key-word sets according to a first query, an intention clarification guidance sentence and a feedback according to an embodiment of the present disclosure
  • FIG. 5 is a flow chart of a method for obtaining one or more key-word sets according to a first query and a feedback according to an embodiment of the present disclosure
  • FIG. 6 is a flow chart of a method for generating a second query according to one or more key-word sets according to an embodiment of the present disclosure
  • FIG. 7 is a block diagram of an interactive searching apparatus according to an embodiment of the present disclosure.
  • FIG. 8 is a block diagram of an interactive searching apparatus according to an embodiment of the present disclosure.
  • FIG. 9 is a block diagram of an interactive searching apparatus according to another embodiment of the present disclosure.
  • FIG. 10 is a block diagram of a second obtaining sub-module in a generating module of an interactive searching apparatus according to an embodiment of the present disclosure
  • FIG. 11 is a block diagram of a third obtaining sub-module in a generating module of an interactive searching apparatus according to an embodiment of the present disclosure
  • FIG. 12 is a block diagram of a generating sub-module in a generating module of an interactive searching apparatus according to an embodiment of the present disclosure.
  • FIG. 13 is a block diagram of an apparatus according to an embodiment of the present disclosure.
  • an interactive search is a search in which a search guidance for a user can be performed by providing an interactive information.
  • the interactive information (such as an intention clarification guidance sentence) is provided for the user according to a query of the user and the query is updated according to a feedback corresponding to the interactive information, such that another search is performed according to the updated query and another search result is returned.
  • embodiments of the present disclosure provide an interactive searching method, including: receiving a first query; obtaining an intention clarification guidance sentence according to the first query; receiving a feedback corresponding to the intention clarification guidance sentence and generating a second query according to the first query, the intention clarification guidance sentence and the feedback; and providing a search result according to the second query.
  • FIG. 1 is a flow chart of an interactive searching method according to an embodiment of the present disclosure. As shown in FIG. 1 , the interactive searching method includes following steps.
  • step S 101 a first query is received.
  • the first query may be a term or a sentence.
  • the user can input the first query in a search box or other search fields in a search page according to a requirement.
  • step S 102 an intention clarification guidance sentence is obtained according to the first query.
  • a sentence library shall be established, such that the search engine can obtain the intention clarification guidance sentence by querying the sentence library.
  • the search engine may obtain one or more intention clarification guidance sentences according to the first query so as to guide the user to clarify a search intention of his own. For example, for the first query such as “What universities can a student be admitted to with a score of 610 points?”, two intention clarification guidance sentences such as “Where are you from?” and “are you a liberal art student or a science student?” can be obtained.
  • a feedback corresponding to the intention clarification guidance sentence is received and a second query is generated according to the first query, the intention clarification guidance sentence and the feedback.
  • the feedback is an answer term or sentence corresponding to the intention clarification guidance sentence.
  • the search engine can receive the feedback corresponding to the intention clarification guidance sentence via a client, in which the feedback is input by the user.
  • the search engine can display the intention clarification guidance sentence to the user via the client and provide at least one candidate result corresponding to the intention clarification guidance sentence or an input box, such that the user can select a candidate result that satisfies his search intention from the at least one candidate result or input an answer corresponding to the intention clarification guidance sentence in the input box directly.
  • the search engine obtains the second query according to the first query, the intention clarification guidance sentence and the feedback. Specifically, the search engine obtains terms indicating the search intention of the user from the first query, the intention clarification guidance sentence and the feedback via a syntax analysis to analyze a structure of a sentence or a word analysis to analyze a meaning of a word, and obtains the second query according to the terms.
  • the search engine provides two intention clarification guidance sentences such as “Where are you from?” and “are you a liberal art student or a science student?” to the user via the client.
  • the input box is provided after each intention clarification guidance sentence, and in this way the user can input “Shandong” and “science” in the two input boxes respectively.
  • the search engine receives the feedback corresponding to the two intention clarification guidance sentences and generates the second query that “What universities can a science student in Shandong province be admitted to with a score of 610 points?”.
  • a search result is provided according to the second query.
  • the search engine after the search result is obtained according to the second query, the search engine returns the search result to the client so as to provide the search result to the user via the client.
  • the search engine may determine whether to further guide the user according to the generated second query; if yes, a corresponding intention clarification guidance sentence may be obtained according to the second query to continue to guide the user; if not, the search result may be provided directly.
  • the search engine provides the search result as shown in FIG. 2 b according to the second query that “What universities can a science student in Shandong province be admitted to with a score of 610 points?” and further provides an intention clarification guidance sentence “What major?”.
  • the search engine after the intention clarification guidance sentence is obtained according to the first query, the search engine provides the at least one candidate result corresponding to the intention clarification guidance sentence, receives a triggering operation for the at least one candidate result and treats a triggered candidate result as the feedback corresponding to the intention clarification guidance sentence, and thus the input operation of the user is reduced.
  • the search engine provides four candidate results such as “scientific research”, “teaching”, “employment” and “dormitory” for the user to select.
  • the input box is further provided, and the user can input the feedback therein if there is not a result satisfying the requirement of himself in the candidate results. Furthermore, when the user clicks “employment”, the search engine generates the second query that “How about the employment of Harbin institute of technology?” and provides the search result as shown in FIG. 2 d.
  • the second query may be generated by selecting a whole or a part of the first query, the intention clarification guidance sentence and the feedback according to a type of the intention clarification guidance sentence.
  • the first query is represented as c_query
  • the intention clarification guidance sentence is represented as qb
  • the feedback is represented as ans
  • the second query is represented as n_query.
  • the type of the intention clarification guidance sentence generally includes a general question, a special question and an alternative question.
  • the second queries generated according to the above three types of intention clarification guidance sentences are shown in Table 1.
  • FIG. 3 is a flow chart of a method for generating a second query according to a first query, an intention clarification guidance sentence and a feedback includes following steps. As shown in FIG. 3 , the method includes following steps.
  • step S 301 a type of the intention clarification guidance sentence is obtained.
  • the type of the intention clarification guidance sentence may be obtained by performing the syntax analysis thereon.
  • step S 302 if the intention clarification guidance sentence is a general question or a special question, one or more key-word sets are obtained according to the first query, the intention clarification guidance sentence and the feedback.
  • a method for obtaining the one or more key-word sets according to the first query, the intention clarification guidance sentence and the feedback includes following steps.
  • the first query, the intention clarification guidance sentence and the feedback are segmented into words so as to obtain a first set of words corresponding to the first query, a second set of words corresponding to the intention clarification guidance sentence and a third set of words corresponding to the feedback.
  • the intention clarification guidance sentence and the feedback into words simultaneously the syntax analysis, an entity identification and a word deletion to delete a word which is not allowed to be used are performed on the first query, the intention clarification guidance sentence and the feedback and a part-of-speech tagging to tag a property of a word is performed on each word, such that the first set of words, the second set of words and the third set of words can be obtained.
  • step S 402 a plurality of features of each word in the first set of words, the second set of words and the third set of words are obtained respectively, and feature values of the plurality of feature of each word are obtained.
  • the plurality of features of each word may include a part of speech, a syntactic constituent, a word frequency in a corpus of a large number of sentences, a number of occurrence times, whether being an entity or not, a position in the sentence and features of a hypernym and a hyponym of the each word.
  • a score of each word is obtained according to the feature values of the plurality of features of each word.
  • the score of each word may be obtained according to the feature values of the plurality of features of each word by using following formulas:
  • score w is a score of w th word
  • ⁇ i is a weight of a i th feature of the w th word
  • f i (w) is a feature value of the i th feature of the w th word
  • N is a total number of the plurality of features of the w th word.
  • a first key-word set is selected from the first set of words
  • a second key-word set is selected from the second set of words
  • a third key-word set is selected from the third set of words according to the score of each word.
  • the key-word set may be selected from each set of words according to a predetermined selecting rule which is not limited herein. For example, a predetermined number of key words having a high score may be selected from each set of words, or the key words having a score higher than a predetermined threshold may be selected.
  • step S 303 if the intention clarification guidance sentence is an alternative question, the one or more key-word sets are obtained according to the first query and the feedback.
  • a method for obtaining the one or more key-word sets according to the first query and the feedback includes following steps.
  • the first query and the feedback are segmented into words so as to obtain a fourth set of words corresponding to the first query and a fifth set of words corresponding to the feedback.
  • the syntax analysis, the entity identification and the word deletion are performed on the first query and the feedback and a part-of-speech tagging is performed on each word, such that the fourth set of words and the fifth set of words can be obtained.
  • step S 502 a plurality of features of each word in the fourth set of words and the fifth set of words are obtained respectively, and feature values of the plurality of features of each word are obtained.
  • the plurality of features of each word may include the part of speech, the syntactic constituent, the word frequency in the corpus of the large number of sentences, the number of occurrence times, whether being the entity or not, the position in the sentence and features of the hypernym and the hyponym of the each word.
  • a score of each word is obtained according to the feature values of the plurality of features of each word.
  • the score of each word may be obtained according to the feature values of the plurality of features of each word by using following formulas:
  • score w is a score of w th word
  • ⁇ i is a weight of a i th feature of the w th word
  • f i (w) is a feature value of the i th feature of the w th word
  • N is a total number of the plurality of features of the w th word.
  • a fourth key-word set is selected from the fourth set of words and a fifth key-word set is selected from the fifth set of words according to the score of each word.
  • the key-word set may be selected from each set of words according to the predetermined selecting rule which is not limited herein.
  • the predetermined number of key words having a high score may be selected from each set of words, or the key words having a score higher than the predetermined threshold may be selected.
  • the second query is generated according to the one or more key-word sets.
  • a method for generating the second query according to the one or more key-word sets includes following steps.
  • a synonym processing is performed on the one or more key-word sets to obtain one or more key-word sequences.
  • a main part of the second query is the first query, and the key words of the intention clarification guidance sentence and the feedback are configured as a supplement part of the second query.
  • the synonym processing may be performed on the one or more key-word sets to obtain the one or more key-word sequences.
  • step S 602 the key words in each of the one or more key-word sequences are sequenced to obtain a plurality of candidate sequences.
  • a complexity of three key-word sequences having x key words, y key words and z key words respectively is x*y*z. Therefore, if there are a large number of key words in the key-word sequence, it is extremely complex to obtain all the possible candidate sequences and a huge calculated amount is needed.
  • the plurality of candidate sequences may be searched and enumerated by a pruning algorithm which is not limited herein, for example the pruning algorithm may be Beam-search and A*.
  • a score for each candidate sequence is obtained according to a sequence and features of the key words in each candidate sequence.
  • the score of each of the plurality of candidate sequences may be obtained by a following formula:
  • score sen is the score of the candidate sequence sen
  • c(w i w i-1 w i-2 ) is a number of times that the key words w i , w i-1 , w i-2 appear in the corpus at a same time
  • c(w i-1 w i-2 ) is a number of times that the key words w i-1 , w i-2 appear in the corpus at a same time
  • l(w i ) ⁇
  • is a sequencing penalty term
  • ⁇ is a constant which is larger than zero and less than one
  • d pos ori — str (w i ) ⁇ pos c — str (w i ), pos ori — str (w i ) is a relative position in a key-word sequence in which the key word w i is between the key word w i and other key words in the key-word sequence in which the key word w i is,
  • ⁇ 1 .
  • the second query is selected from the plurality of candidate sequences according to the score of each of the plurality of candidate sequences.
  • the candidate sequence having a highest score may be selected from the plurality of candidate sequences as the second query.
  • the intention clarification guidance sentence provided by the search engine is obtained, and the query is updated according to the intention clarification guidance sentence and the feedback corresponding to the intention clarification guidance sentence and the search result is provided according to the updated query, such that the search intention of the user is clarified, and also the query can be updated automatically according to a historical query and the feedback and it is just required for the user to input the feedback corresponding to the intention clarification guidance sentence, and thus an input operation of the user is reduced and a memory burden of the user is also decreased.
  • an accuracy of the search engine to identify the search intention of the user is increased and a requirement of the user is satisfied, and in this way a user experience is improved.
  • an interactive searching apparatus is provided by embodiments of the present disclosure.
  • the interactive searching apparatus includes a first receiving module configured to receive a first query; a first obtaining module configured to obtain an intention clarification guidance sentence according to the first query; and a second receiving module configured to receive a feedback corresponding to the intention clarification guidance sentence; a generating module configured to generate a second query according to the first query, the intention clarification guidance sentence and the feedback; and a providing module configured to provide a search result according to the second query.
  • FIG. 7 is a block diagram of an interactive searching apparatus according to an embodiment of the present disclosure.
  • the interactive searching apparatus includes a first receiving module 100 , a first obtaining module 200 , a second receiving module 300 , a generating module 400 and a providing module 500 .
  • the first receiving module 100 is configured to receive a first query.
  • the first query may be a term or a sentence.
  • a user can input the first query in a search box or other search fields in a search page according to a requirement.
  • the first obtaining module 200 is configured to obtain an intention clarification guidance sentence according to the first query.
  • a sentence library shall be established, such that the first obtaining module 200 can obtain the intention clarification guidance sentence by querying the sentence library.
  • a step of establishing the sentence library can be omitted.
  • the first obtaining module 200 may obtain one or more intention clarification guidance sentences according to the first query so as to guide the user to clarify a search intention of his own.
  • the first obtaining module 200 For example, for the first query such as “What universities can a student be admitted to with a score of 610 points?”, two intention clarification guidance sentences such as “Where are you from?” and “are you a liberal art student or a science student?” can be obtained by the first obtaining module 200 .
  • the second receiving module 300 is configured to receive a feedback corresponding to the intention clarification guidance sentence.
  • the feedback is an answer term or sentence corresponding to the intention clarification guidance sentence.
  • the second receiving module 300 can receive the feedback corresponding to the intention clarification guidance sentence via a client, in which the feedback is input by the user.
  • the intention clarification guidance sentence may be displayed to the user via the client and at least one candidate result corresponding to the intention clarification guidance sentence or an input box may be provided, such that the user can select a candidate result that satisfies his search intention from the at least one candidate result or input an answer corresponding to the intention clarification guidance sentence in the input box directly.
  • the generating module 400 is configured to generate a second query according to the first query, the intention clarification guidance sentence and the feedback. Specifically, the generating module 400 obtains terms indicating the search intention of the user from the first query, the intention clarification guidance sentence and the feedback via a syntax analysis to analyze a structure of a sentence or a word analysis to analyze a meaning of a word, and obtains the second query according to the terms.
  • the generating module 400 generates the second query by selecting a whole or a part of the first query, the intention clarification guidance sentence and the feedback according to a type of the intention clarification guidance sentence.
  • the first query is represented as c_query
  • the intention clarification guidance sentence is represented as qb
  • the feedback is represented as ans
  • the second query is represented as n_query.
  • the type of the intention clarification guidance sentence generally includes a general question, a special question and an alternative question.
  • the second queries generated according to the above three types of intention clarification guidance sentences are shown in Table 1.
  • two intention clarification guidance sentences such as “Where are you from?” and “are you a liberal art student or a science student?” may be provided to the user via the client.
  • the input box is provided after each intention clarification guidance sentence, and in this way the user can input “Shandong” and “science” in the two input boxes respectively.
  • the second receiving module 300 receives the feedback corresponding to the two intention clarification guidance sentence and the generating module 400 generates the second query that “What universities can a science student in Shandong province be admitted to with a score of 610 points?”.
  • the providing module 500 is configured to provide a search result according to the second query.
  • the search result may be returned to the client so as to be provided to the user via the client.
  • the providing module 500 provides the search result as shown in FIG. 2 b according to the second query that “What universities can a science student in Shandong province be admitted to with a score of 610 points?” and further provides an intention clarification guidance sentence “What major?”.
  • the intention clarification guidance sentence provided by the search engine is obtained, and the query is updated according to the intention clarification guidance sentence and the feedback corresponding to the intention clarification guidance sentence and the search result is provided according to the updated query, such that the search intention of the user is clarified, and also the query can be updated automatically according to a historical query and the feedback and it is just required for the user to input the feedback corresponding to the intention clarification guidance sentence, and thus an input operation of the user is reduced and a memory burden of the user is also decreased.
  • an accuracy of the search engine to identify the search intention of the user is increased and a requirement of the user is satisfied, and in this way a user experience is improved.
  • the generating module includes a first obtaining sub-module 410 , a second obtaining sub-module 420 , a third obtaining sub-module 430 and a generating sub-module 440 .
  • the first obtaining sub-module 410 is configured to obtain a type of the intention clarification guidance sentence. Specifically, the first obtaining sub-module 410 obtains the type of the intention clarification guidance sentence by performing a syntax analysis thereon.
  • the second obtaining sub-module 420 is configured to obtain one or more key-word sets according to the first query, the intention clarification guidance sentence and the feedback, if the intention clarification guidance sentence is a general question or a special question.
  • the second obtaining sub-module 420 further includes a first segmenting unit 421 , a first obtaining unit 422 , a second obtaining unit 423 and a first selecting unit 424 .
  • the first segmenting unit 421 is configured to segment the first query, the intention clarification guidance sentence and the feedback into words so as to obtain a first set of words corresponding to the first query, a second set of words corresponding to the intention clarification guidance sentence and a third set of words corresponding to the feedback.
  • the intention clarification guidance sentence and the feedback into words simultaneously the syntax analysis, an entity identification and a word deletion to delete a word which is not allowed to be used are performed on the first query, the intention clarification guidance sentence and the feedback and a part-of-speech tagging to tag a property of a word is performed on each word, such that the first set of words, the second set of words and the third set of words can be obtained.
  • the first obtaining unit 422 is configured to obtain a plurality of features of each word in the first set of words, the second set of words and the third set of words respectively, and to obtain feature values of the plurality of features of each word.
  • the plurality of features of each word may include a part of speed, a syntactic constituent, a word frequency in a corpus of a large number of sentences, a number of occurrence times, whether being an entity or not, a position in the sentence and features of a hypernym and a hyponym of the each word.
  • the second obtaining unit 423 is configured to obtain a score of each word according to the feature values of the plurality of features of each word.
  • the score of each word may be obtained according to the feature values of the plurality of features of each word by using following formulas:
  • score w is a score of w th word
  • ⁇ i is a weight of a i th feature of the w th word
  • f i (w) is a feature value of the i th feature of the w th word
  • N is a total number of the plurality of features of the w th word.
  • the first selecting unit 424 is configured to select a first key-word set from the first set of words, a second key-word set from the second set of words and a third key-word set from the third set of words according to the score of the each word.
  • the key-word set may be selected from each set of words according to a predetermined selecting rule which is not limited herein. For example, a predetermined number of key words having a high score may be selected from each set of words, or the key words having a score higher than a predetermined threshold may be selected.
  • the third obtaining sub-module 430 is configured to obtain the one or more key-word sets according to the first query and the feedback, if the intention clarification guidance sentence is an alternative question.
  • the third obtaining sub-module 430 includes a second segmenting unit 431 , a third obtaining unit 432 , a fourth obtaining unit 433 and a second selecting unit.
  • the second segmenting unit 431 is configured to segment the first query and the feedback into words so as to obtain a fourth set of words corresponding to the first query and a fifth set of words corresponding to the feedback.
  • the syntax analysis, the entity identification and the word deletion are perform on the first query and the feedback and the part-of-speech tagging is performed on each word, such that the fourth set of words and the fifth set of words can be obtained.
  • the third obtaining unit 432 is configured to obtain a plurality of features of each word in the fourth set of words and the fifth set of words respectively, and to obtain feature values of the plurality of features of each word.
  • the plurality of features of each word may include the part of speed, the syntactic constituent, the word frequency in the corpus of the large number of sentences, the number of occurrence times, whether being the entity or not, the position in the sentence and features of the hypernym and the hyponym of the each word.
  • the fourth obtaining unit 433 is configured to obtain a score of each word according to the feature values of the plurality of features of each word.
  • the score of each word may be obtained according to the feature values of the plurality of features of each word by using following formulas:
  • score w is a score of w th word
  • ⁇ i is a weight of a i th feature of the w th word
  • f i (w) is a feature value of the i th feature of the w th word
  • N is a total number of the plurality of features of the w th word.
  • the second selecting unit 434 is configured to select a fourth key-word set from the fourth set of words and a fifth key-word set from the fifth set of words according to the score of each word.
  • the key-word set may be selected from each set of words according to the predetermined selecting rule which is not limited herein. For example, the predetermined number of key words having a high score may be selected from each set of words, or the key words having a score higher than the predetermined threshold may be selected.
  • the generating sub-module 440 is configured to generate the second query according to the one or more key-word sets.
  • the generating sub-module 440 includes a synonym processing 441 , a fifth obtaining unit 442 , a sixth obtaining unit 443 and a third selecting unit 444 .
  • the synonym processing unit 441 is configured to perform a synonym processing on the one or more key-word sets to obtain one or more key-word sequences.
  • a main part of the second query is the first query and the key words of the intention clarification guidance sentence and the feedback are configured as a supplement part of the second query.
  • the synonym processing may be performed on the one or more key-word sets to obtain the one or more key-word sequences.
  • the fifth obtaining unit 442 is configured to sequence the key words in each of the one or more key-word sequences to obtain a plurality of candidate sequences.
  • a complexity of three key-word sequences having x key words, y key words and z key words respectively is x*y*z. Therefore, if there are a large number of key words in the key-word sequence, it is extremely complex to obtain all the possible candidate sequences and a huge calculated amount is needed.
  • the plurality of candidate sequences may be searched and enumerated by a pruning algorithm which is not limited herein, for example the pruning algorithm may be Beam-search and A*.
  • the sixth obtaining unit 443 is configured to obtain a score for each candidate sequence according to a sequence and features of the key words in each candidate sequence.
  • the score of each of the plurality of candidate sequences may be obtained by a following formula:
  • score sen is the score of the candidate sequence sen
  • c(w i w i-1 w i-2 ) is a number of times that the key words w i , w i-1 , w i-2 appear in the corpus at a same time
  • c(w i-1 w i-2 ) is a number of times that the key words w i-1 , w i-2 appear in the corpus at a same time
  • l(w i ) ⁇
  • is a sequencing penalty term
  • ⁇ is a constant which is larger than zero and less than one
  • d pos ori — str (w i ) ⁇ pos c — str (w i ), pos ori — str (w i ) is a relative position in a key-word sequence in which the key word w i is between the key word w i and other key words in the key-word sequence in which the key word w i is,
  • ⁇ 1 .
  • the third selecting unit 444 is configured to select the second query from the plurality of candidate sequences according to the score of the each of the plurality of candidate sequences.
  • the candidate sequence having a highest score may be selected from the plurality of candidate sequences as the second query.
  • FIG. 9 is a block diagram of an interactive searching apparatus according to an embodiment of the present disclosure.
  • the interactive searching apparatus includes: a first receiving module 100 , a first obtaining module 200 , a second receiving module 300 , a generating module 400 , a providing module 500 and a second obtaining module 600 .
  • the second obtaining module 600 is configured to obtain at least one candidate result corresponding to the intention clarification guidance sentence
  • the second obtaining module 600 provides four candidate results such as “scientific research”, “teaching”, “employment” and “dormitory” for the user to select.
  • the input box is further provided, and the user can input the feedback therein if there is not a result satisfying the requirement of himself in the candidate results.
  • the generating module may generates the second query that “How about the employment of Harbin institute of technology”? and the providing module 500 provides the search result as shown in FIG. 2 d.
  • the at least one candidate result corresponding to the intention clarification guidance sentence is provided to the user and the corresponding feedback is obtained according to the triggering operation of the user for the at least one candidate result, such that the second query is obtained and the search result is provided according to the second query.
  • the user it is not needed for the user to input the feedback and the input operation of the user is further reduced.
  • An apparatus 800 is provided according to embodiments of the present disclosure.
  • the apparatus 800 includes: one or more processors 810 ; a memory 820 ; and one or more programs stored in the memory 820 and executed by the one or more processors 810 to execute steps of: receiving a first query; obtaining an intention clarification guidance sentence according to the first query; receiving a feedback corresponding to the intention clarification guidance sentence and generating a second query according to the first query, the intention clarification guidance sentence and the feedback; and providing a search result according to the second query.
  • Apparatus is accessible by a user performing the search.
  • Apparatus 800 may be directly accessible, or remotely accessible through a user device remotely connected to apparatus 800 (e.g. via the internet) such that the query is received by apparatus 800 via the user device; the intention clarification guidance sentence is sent to the user device, and the feedback is received by apparatus 800 from the user device.
  • a non-transitory computer-readable storage medium includes one or more programs for executing steps of: receiving a first query; obtaining an intention clarification guidance sentence according to the first query; receiving a feedback corresponding to the intention clarification guidance sentence and generating a second query according to the first query, the intention clarification guidance sentence and the feedback; and providing a search result according to the second query.
  • Any procedure or method described in the flow charts or described in any other way herein may be understood to include one or more modules, portions or parts for storing executable codes that realize particular logic functions or procedures.
  • advantageous embodiments of the present disclosure includes other implementations in which the order of execution is different from that which is depicted or discussed, including executing functions in a substantially simultaneous manner or in an opposite order according to the related functions. This should be understood by those skilled in the art which embodiments of the present disclosure belong to.
  • the logic and/or step described in other manners herein or shown in the flow chart, for example, a particular sequence table of executable instructions for realizing the logical function may be specifically achieved in any computer readable medium to be used by the instruction execution system, device or equipment (such as the system based on computers, the system including processors or other systems capable of obtaining the instruction from the instruction execution system, device and equipment and executing the instruction), or to be used in combination with the instruction execution system, device and equipment.
  • the computer readable medium may be any device adaptive for including, storing, communicating, propagating or transferring programs to be used by or in combination with the instruction execution system, device or equipment.
  • the computer readable medium include but are not limited to: an electronic connection (an electronic device) with one or more wires, a portable computer enclosure (a magnetic device), a random access memory (RAM), a read only memory (ROM), an erasable programmable read-only memory (EPROM or a flash memory), an optical fiber device and a portable compact disk read-only memory (CDROM).
  • the computer readable medium may even be a paper or other appropriate medium capable of printing programs thereon, this is because, for example, the paper or other appropriate medium may be optically scanned and then edited, decrypted or processed with other appropriate methods when necessary to obtain the programs in a electric manner, and then the programs may be stored in the computer memories.
  • each part of the present disclosure may be realized by the hardware, software, firmware or their combination.
  • a plurality of steps or methods may be realized by the software or firmware stored in the memory and executed by the appropriate instruction execution system.
  • the steps or methods may be realized by one or a combination of the following techniques known in the art: a discrete logic circuit having a logic gate circuit for realizing a logic function of a data signal, an application-specific integrated circuit having an appropriate combination logic gate circuit, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.
  • each function cell of the embodiments of the present disclosure may be integrated in a processing module, or these cells may be separate physical existence, or two or more cells are integrated in a processing module.
  • the integrated module may be realized in a form of hardware or in a form of software function modules. When the integrated module is realized in a form of software function module and is sold or used as a standalone product, the integrated module may be stored in a computer readable storage medium.
  • the storage medium mentioned above may be read-only memories, magnetic disks or CD, etc.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Computational Linguistics (AREA)
  • Mathematical Physics (AREA)
  • Human Computer Interaction (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
US14/572,393 2014-05-27 2014-12-16 Interactive searching method and apparatus Abandoned US20150347500A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201410228820.6 2014-05-27
CN201410228820.6A CN103995880B (zh) 2014-05-27 2014-05-27 交互式搜索方法和装置

Publications (1)

Publication Number Publication Date
US20150347500A1 true US20150347500A1 (en) 2015-12-03

Family

ID=51310045

Family Applications (1)

Application Number Title Priority Date Filing Date
US14/572,393 Abandoned US20150347500A1 (en) 2014-05-27 2014-12-16 Interactive searching method and apparatus

Country Status (4)

Country Link
US (1) US20150347500A1 (enrdf_load_stackoverflow)
EP (1) EP2953038A1 (enrdf_load_stackoverflow)
JP (1) JP5998194B2 (enrdf_load_stackoverflow)
CN (1) CN103995880B (enrdf_load_stackoverflow)

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20180157721A1 (en) * 2016-12-06 2018-06-07 Sap Se Digital assistant query intent recommendation generation
CN109284405A (zh) * 2018-08-31 2019-01-29 北京优酷科技有限公司 信息应答方法及装置
CN109766414A (zh) * 2019-01-18 2019-05-17 广东小天才科技有限公司 一种意图识别方法及系统
KR102144370B1 (ko) * 2019-11-18 2020-08-13 주식회사 오투오 대화형 정보 검색장치
CN111930904A (zh) * 2020-07-08 2020-11-13 联想(北京)有限公司 信息应答方法、装置、设备及存储介质
US20210081500A1 (en) * 2019-09-18 2021-03-18 International Business Machines Corporation Hypernym detection using strict partial order networks

Families Citing this family (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104156492A (zh) * 2014-09-02 2014-11-19 北京国双科技有限公司 搜索内容的提示方法和装置
CN104571813B (zh) * 2014-12-12 2019-03-29 百度在线网络技术（北京）有限公司 一种信息的显示方法及装置
CN106653006B (zh) * 2016-11-17 2019-11-08 百度在线网络技术（北京）有限公司 基于语音交互的搜索方法和装置
CN106681598B (zh) * 2017-01-13 2020-12-15 北京百度网讯科技有限公司 信息输入方法和装置
CN107168987A (zh) * 2017-03-24 2017-09-15 联想(北京)有限公司 一种数据处理方法及其装置
CN107133280A (zh) * 2017-04-14 2017-09-05 合信息技术（北京）有限公司 反馈的响应方法及装置
CN108304434B (zh) * 2017-09-04 2021-11-05 腾讯科技（深圳）有限公司 信息反馈方法和终端设备
CN110309274B (zh) * 2018-03-14 2021-09-07 北京三快在线科技有限公司 引导语推荐方法、装置及电子设备
CN109902149B (zh) * 2019-02-21 2021-08-13 北京百度网讯科技有限公司 查询处理方法和装置、计算机可读介质
CN112162955A (zh) * 2020-10-29 2021-01-01 海信视像科技股份有限公司 用户日志的处理装置及方法
CN114627864A (zh) * 2020-12-10 2022-06-14 海信视像科技股份有限公司 显示设备与语音交互方法
CN119621254B (zh) * 2025-02-13 2025-05-23 招商局国际科技有限公司 多线程并发的澄清与多轮对话逻辑处理方法、装置及设备

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20010049688A1 (en) * 2000-03-06 2001-12-06 Raya Fratkina System and method for providing an intelligent multi-step dialog with a user
US20120023119A1 (en) * 2009-03-30 2012-01-26 Ducatel Gery M Data searching system
US20120303356A1 (en) * 2011-05-27 2012-11-29 International Business Machines Corporation Automated self-service user support based on ontology analysis
US20130018895A1 (en) * 2011-07-12 2013-01-17 Harless William G Systems and methods for extracting meaning from speech-to-text data

Family Cites Families (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP4003468B2 (ja) * 2002-02-05 2007-11-07 株式会社日立製作所 適合性フィードバックによる類似データ検索方法および装置
JP4650072B2 (ja) * 2005-04-12 2011-03-16 富士ゼロックス株式会社 質問応答システム、およびデータ検索方法、並びにコンピュータ・プログラム
CN100421113C (zh) * 2006-03-03 2008-09-24 中国移动通信集团公司 基于个性化信息的搜索系统及搜索方法
JP5246932B2 (ja) * 2008-08-29 2013-07-24 西日本電信電話株式会社 検索装置及び方法、ならびに、コンピュータプログラム
JP4795452B2 (ja) * 2009-04-30 2011-10-19 沖電気工業株式会社 検索システム及び検索プログラム
CN101937437B (zh) * 2009-06-30 2011-11-16 华为技术有限公司 一种搜索方法、装置和系统
KR101734970B1 (ko) * 2010-02-10 2017-05-12 오의진 사용자 검색의도에 부합하는 검색 결과 제공 방법 및 시스템
US8280900B2 (en) * 2010-08-19 2012-10-02 Fuji Xerox Co., Ltd. Speculative query expansion for relevance feedback
CN102456018B (zh) * 2010-10-18 2016-03-02 腾讯科技（深圳）有限公司 一种交互搜索方法及装置
JP2012248161A (ja) * 2011-05-31 2012-12-13 Oki Electric Ind Co Ltd 対話型検索システム及びプログラム、並びに、対話シナリオ生成システム及びプログラム
US9767144B2 (en) * 2012-04-20 2017-09-19 Microsoft Technology Licensing, Llc Search system with query refinement
JP5880350B2 (ja) * 2012-08-24 2016-03-09 富士ゼロックス株式会社 情報検索プログラム及び情報検索装置

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20010049688A1 (en) * 2000-03-06 2001-12-06 Raya Fratkina System and method for providing an intelligent multi-step dialog with a user
US20120023119A1 (en) * 2009-03-30 2012-01-26 Ducatel Gery M Data searching system
US20120303356A1 (en) * 2011-05-27 2012-11-29 International Business Machines Corporation Automated self-service user support based on ontology analysis
US20130018895A1 (en) * 2011-07-12 2013-01-17 Harless William G Systems and methods for extracting meaning from speech-to-text data

Cited By (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10866975B2 (en) 2016-12-06 2020-12-15 Sap Se Dialog system for transitioning between state diagrams
US10503744B2 (en) 2016-12-06 2019-12-10 Sap Se Dialog system for transitioning between state diagrams
US11314792B2 (en) * 2016-12-06 2022-04-26 Sap Se Digital assistant query intent recommendation generation
US10810238B2 (en) 2016-12-06 2020-10-20 Sap Se Decoupled architecture for query response generation
US20180157721A1 (en) * 2016-12-06 2018-06-07 Sap Se Digital assistant query intent recommendation generation
CN109284405A (zh) * 2018-08-31 2019-01-29 北京优酷科技有限公司 信息应答方法及装置
CN109766414A (zh) * 2019-01-18 2019-05-17 广东小天才科技有限公司 一种意图识别方法及系统
US11068665B2 (en) * 2019-09-18 2021-07-20 International Business Machines Corporation Hypernym detection using strict partial order networks
US20210081500A1 (en) * 2019-09-18 2021-03-18 International Business Machines Corporation Hypernym detection using strict partial order networks
US20210303800A1 (en) * 2019-09-18 2021-09-30 International Business Machines Corporation Hypernym detection using strict partial order networks
US11694035B2 (en) * 2019-09-18 2023-07-04 International Business Machines Corporation Hypernym detection using strict partial order networks
KR102144370B1 (ko) * 2019-11-18 2020-08-13 주식회사 오투오 대화형 정보 검색장치
CN111930904A (zh) * 2020-07-08 2020-11-13 联想(北京)有限公司 信息应答方法、装置、设备及存储介质

Also Published As

Publication number Publication date
CN103995880A (zh) 2014-08-20
EP2953038A1 (en) 2015-12-09
CN103995880B (zh) 2019-03-12
JP5998194B2 (ja) 2016-09-28
JP2015225657A (ja) 2015-12-14

Similar Documents

Publication Publication Date Title
US20150347500A1 (en) Interactive searching method and apparatus
US11016966B2 (en) Semantic analysis-based query result retrieval for natural language procedural queries
US10606915B2 (en) Answer searching method and device based on deep question and answer
US10997503B2 (en) Computationally efficient neural network architecture search
CN107491547B (zh) 基于人工智能的搜索方法和装置
CN109815487B (zh) 文本质检方法、电子装置、计算机设备及存储介质
US20150339385A1 (en) Interactive searching method and apparatus
CN116911312B (zh) 一种任务型对话系统及其实现方法
US9785672B2 (en) Information searching method and device
US12008473B2 (en) Augmenting machine learning language models using search engine results
US20150293978A1 (en) Interactive searching and recommanding method and apparatus
US10055453B2 (en) Interactive searching method and apparatus
US10275454B2 (en) Identifying salient terms for passage justification in a question answering system
US9613133B2 (en) Context based passage retrieval and scoring in a question answering system
CN109947952B (zh) 基于英语知识图谱的检索方法、装置、设备及存储介质
US10032448B1 (en) Domain terminology expansion by sensitivity
CN113761868B (zh) 文本处理方法、装置、电子设备及可读存储介质
US20120209590A1 (en) Translated sentence quality estimation
CN109524008A (zh) 一种语音识别方法、装置及设备
AU2023236937A1 (en) Generating output sequences with inline evidence using language model neural networks
JP2024174994A (ja) ビデオ生成および編成モデル取得方法、装置、デバイスおよび記憶媒体
CN117150044A (zh) 基于知识图谱的专利处理方法、装置及存储介质
CN117494815A (zh) 面向档案的可信大语言模型训练、推理方法和装置
CN118246551A (zh) 垂直领域模型推理加速方法、装置
CN112559711A (zh) 一种同义文本提示方法、装置及电子设备

Legal Events

Date Code Title Description
AS Assignment

Owner name: BAIDU ONLINE NETWORK TECHNOLOGY (BEIJING) CO., LTD

Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:LI, TINGTING;WAN, WEI;ZHAO, SHIQI;REEL/FRAME:034741/0232

Effective date: 20150106

STCB Information on status: application discontinuation

Free format text: ABANDONED -- FAILURE TO RESPOND TO AN OFFICE ACTION